4

Communicating Science in a Complex, Competitive Communication Environment

Recent years have seen widespread and dramatic changes in the way people seek or encounter information about science and the issues of concern to them. These changes present additional challenges to communicators of science regardless of whether the science is involved in public controversy. Gone are the days when citizens obtained much of their information from relatively few radio and television networks, magazines, and newspapers. Today, people may encounter scientific information from a wide variety of media operations, as well as blogs, social media feeds, podcasts, YouTube channels, and a host of other sources that did not exist 20 years ago. At the same time, some of the ways citizens have in the past received high-quality information about science and social issues are fading away. Traditional news sources, for example, face severe economic challenges. Newspaper circulation continues to decline with each passing year (Pew Research Center, 2016a), reducing the number of professional journalists writing news. American newspapers in 2014 had some 33,000 full-time newsroom employees—20,000 fewer than they did in 1994 (Pew Research Center, 2016a). This chapter examines in turn trends in the communication of science news, how journalistic decisions affect the coverage of and audiences for science, and emerging research on use of the Internet as a source of science news.

TRENDS IN THE COMMUNICATION OF SCIENCE NEWS

While the ongoing changes summarized above affect the media system more broadly, they have had particularly strong impacts on how science

is communicated in the media. Three trends are particular noteworthy (Scheufele, 2013).

First, as noted above, people have shifted away from traditional media (especially newspapers and television) and toward online-supplemented or online-only news for scientific information. This trend is especially pronounced among younger and scientifically literate audiences (Su et al., 2015). In 2014, 47 percent of Americans reported that the Internet was their primary source of news and information about science and technology, up from just 9 percent in 2001. In comparison, 28 percent said television was their primary source of such information, down from 32 percent in 2012. As for other sources, 7 percent named print magazines as their top source, 6 percent print newspapers, and 3 percent radio (National Science Board, 2016a). Younger, higher-earning, and better-educated Americans were more likely to say they receive most of their information about science and technology by way of the Internet, online newspapers, online magazines, or similar sources. Older and less-educated Americans were more likely to continue to rely on television news and print sources (National Science Board, 2016a).

Of the roughly half of Americans who said they rely primarily on the Internet for information about science and technology, a little more than a third said they use a search engine such as Google; a combined 45 percent said they use primarily online newspapers (23 percent), online magazines (15 percent), or other online news sites (7 percent). Just 8 percent of Internet information seekers—or 3 percent of all Americans—said they rely on a science-focused site as their primary source of information about science and technology (National Science Board, 2016a). Similarly, in a 2015 poll conducted by the Associated Press and other organizations that asked people their top three sources of information about science and technology, more than 50 percent of American adults identified search engines, just over 40 percent cited Facebook, and more than 30 percent said conversations with friends and family (Brossard, 2016). Several tools available through Google enable keyword searches and queries that provide a potentially useful source of data about the public’s interest in science generally and in specific topics over time, including how public interests compare with news coverage. However, such tools are difficult to apply across countries and do not account for the interests of those without access to the Internet (Baram-Tsabari and Segev, 2009; Segev and Baram-Tsabari, 2010).

Online information environments—especially since the advent of smartphones and other mobile devices—have the potential to deliver more scientific information more quickly than ever before and make their users largely independent of physical repositories, such as libraries (Scheufele and Nisbet, 2012). At the same time, web 2.0 environments (websites that encourage and depend upon content created by their visitors) have made it

possible for users to debate news among themselves and to comment on, share, and repurpose information (to some degree) independently of commercial media outlets.

The second noteworthy trend is that “news holes”—the number of column inches devoted to news in print or the time available for news on television—have been shrinking. For science coverage, this trend has been particularly dramatic. In 1989, 95 U.S. newspapers had weekly science sections; by 2012, only 19 still did (Morrison, 2013).

As might be expected, this decline is an important driver of the third key trend: the shrinking number of science journalists employed full-time at news outlets. Professional journalists with knowledge of the science about which they are reporting often have a different but important role to play in people’s understanding of science,¹ but represent a smaller proportion of the science information available to media consumers even as the scope and size of that information have increased (Scheufele, 2013).

Although the traditional infrastructure for science news is declining, communicators have new ways to participate in public debates about science, and more scientists than ever are now speaking directly to the public via blogs, podcasts, YouTube videos, and the like. This widening of channels for scientific information has allowed more science communicators to try to reach the public. But it also means there are more actors in the media landscape who may, either intentionally or unintentionally, provide inaccurate science information. So while today’s science media landscape is likely larger than the declining mass media/newspaper-delivery system of the past, that expanding landscape does not offer clear mechanisms for filtering out false, sensational, and misleading information. More than ever before, citizens are left to their own devices as they struggle to determine whom to trust and what to believe about science-related controversies.

This is the new, and not entirely understood, media environment with which science communicators must cope.

HOW JOURNALISTIC DECISIONS AFFECT SCIENCE COVERAGE AND AUDIENCES

Despite the growing impact of new media, much of the scientific information Americans receive through media still originates from traditional journalism, including information transmitted via links on social media. Therefore, science communicators need to understand the tools used by journalists to shape scientific information, especially that related to contentious societal issues.

___________________

¹ For further discussion of journalistic goals and norms versus those of science, see Nelkin (1996) and Dunwoody (2014).

How Science Is Covered in Mainstream News

A number of factors, closely related to framing (see Chapter 2), influence journalists’ decisions about how much attention contentious issues and the related science receive (agenda building) and how those issues and the science are defined or portrayed (for an overview, see Scheufele, 1999). Among the most important of these factors are the tactics journalists can use to present new information. Journalists covering complex policy debates tend to rely on a few standardized plots and news formats (Bennett, 2016). When a new issue, such as food biotechnology or stem cell research, taps familiar themes from previous political conflicts or evokes story lines familiar from popular culture and history, journalists cast the new issue’s actors and events to fit these well-known story lines.

Science-related policy debates become most newsworthy, for example, when they involve clear conflict leading to a promise of resolution, such as the passage of legislation, the veto of a bill, or an international climate change summit—a scenario that Cook (1996) describes as “conflict with movement.” When policy making appears to be stuck in a quagmire, with no movement on legislation or no political decision on the horizon, news coverage of a complex science-related issue cannot build anticipation for a resolution. As a result, news reporting on such issues is less frequent and less successful in garnering attention (Nisbet and Huge, 2006, 2007).

Similarly, once a conflict appears to have been resolved, even if only partially or temporarily, the inherent drama of the issue is lost for many journalists. Absent a new scientific discovery or the need for a political decision on the horizon, journalists need a drama on which to focus (McComas and Shanahan, 1999). This need for effective narrative explains why the perceived political relevance of a science-related issue is critically important to how much news coverage it receives. On most science-related issues, journalists provide a relatively steady amount of coverage as they follow the release of each new study on an issue. But when an issue such as biotechnology or climate change receives attention because of a congressional debate or White House statement, that issue becomes more relevant to political conflict. The issue then gains the attention of political journalists, general assignment reporters, and opinion writers who do not regularly communicate about science. These journalists outnumber science writers, and political conflict receives more coverage than science. Thus when an issue becomes defined as politically relevant, the potential for it to receive a high volume of coverage increases (Kepplinger, 1992, 1995; Nisbet and Huge, 2006, 2007).

Of course, experienced news sources formulate their message strategies to accent familiar themes (Nisbet and Huge, 2006, 2007; Nisbet et al., 2003). Hence, journalists often evoke images of ethical quandaries or

extremes of risk or benefit to meet the need for dramatic structure. Meanwhile, political reporters, who tend to frame almost all aspects of public life as if they were simply a matter of political competition (Cappella and Jamieson, 1997; Patterson, 1994), impose their own interpretations on science stories. Their framing of an issue emphasizes who is ahead or behind and the episodic day-to-day tactics employed by strategic actors to gain an advantage. These frames generate drama, first by focusing on conflict, with an expectation of winners and losers, and second by personalizing the news through a focus on individual battles between specific actors. This framing spurs political journalists to treat opposing claims as if they were equal—leaving the public with the impression, for example, that scientific debate about climate change has equal numbers of advocates on both sides (a phenomenon known as “false balance”; see Chapter 3) (Bennett, 2016; Nisbet et al., 2003).

It is important to understand and track over time how science is covered in the media because, as discussed in the next section, coverage can affect audience perceptions of the issues that should be of concern to them personally and should receive priority in public policy agendas. Moreover, media coverage can be expected to affect how science communication from other sources is interpreted, a contextual factor that needs to be considered in research on science communication.

How Coverage of Science Affects Public Perceptions

By giving attention to some issues over others, the media influence what the public perceives as most pressing and important (Iyengar and Kinder, 1987; McCombs, 2004). Consistent with this hypothesis, studies tracking news attention and public perceptions typically find that a rise in overall news attention to an issue precedes a rise in public concern (McCombs, 2004; Rogers et al., 1991). In addition, laboratory experiments indicate that when subjects are repeatedly shown newscasts over a week, the top issues featured in those newscasts emerge as among the subjects’ primary national concerns (see Iyengar and McGrady, 2007, for discussion). When individuals are asked to describe the issues that are of most concern to them or are the most important facing the country, or to reflect on how worried they are about a risk, their responses are most likely to reflect the extent of their media exposure to the issues, as well as whether the issues affect them directly (Kim et al., 2002; Scheufele, 2000). Recent experience with Ebola and swine flu reflect these effects (Pew Research Center, 2009b, 2014). This is why the narratives that shape science coverage in the media are important to science communication. The inappropriate fears over Ebola and swine flu fanned by the media illustrate how the media’s narrative frames can make it difficult to get accurate science onto the public agenda.

The amount of coverage received by issues or events in the news also often serves as the criterion by which the public evaluates the performance and credibility of a political leader, government agency, scientific organization, or corporation (Iyengar and Kinder, 1987; Nisbet and Feldman, 2011). Political leaders and organizations recognize and anticipate these effects. As a result, when an issue or event rises on the overall news agenda and thereby becomes an object of public concern, political actors are likely to take some form of action to address it, even if only symbolically. This action may be taken with little to no expert advice on the subject given that, as discussed in Chapter 3, actions on contentious issues in society often are driven by values, beliefs, and other considerations.

Thus research on the agenda-setting effect of the media has repeatedly provided evidence that the issues portrayed in the media shape the issue priorities of the public. The effects of such agenda setting can be both positive and negative. In the case of Ebola, for example, news coverage that drove the issue to the top of the public’s agenda led policy makers to call for restricting air travel from Africa and quarantining all medical personnel returning from the area, both of which ran counter to the advice of the Centers for Disease Control and Prevention. At the same time, increased coverage of the epidemic led policy makers to make a substantial investment in controlling it. The effect of media coverage on the priorities of policy-making institutions in general, however, is less well understood than its effect on the priorities of the public. Research in this area has yielded mixed results (Kingdon, 1995), and whether the media drive policy or the reverse is unclear (Wolfe et al., 2013).

More is known about journalistic decisions and their effects on audiences in traditional news outlets than in new media. An important question for research is how these processes operate and affect audiences for scientific information in rapidly changing online environments. As their audiences have decreased and turned to online media as primary sources for scientific information, traditional media organizations have devoted less space and time to science (National Science Board, 2016a; Su et al., 2015). There is as yet no clearly viable business model that can sustain traditional journalistic organizations (and thus their practices) in an online environment.

On the one hand, online channels have made information more easily and more widely accessible. On the other hand, new information-filtering tools that people can use to block information with which they disagree have made it easier than ever to hear and see only news that accords with one’s beliefs (Scheufele and Nisbet, 2012). Thus the online information environment has been called an “echo chamber” (Sunstein, 2009) and a “filter bubble” (Pariser, 2012). As noted previously, all people filter information

through preexisting beliefs. At the dawn of web 2.0, some were optimistic that an Internet full of user-controlled web experiences and user-generated content would become a platform for informative and helpful dialogue about science. This view has since been tempered by the emergence of filter bubbles, abundant uncivil online comments on science news stories, and the detrimental effects of those comments on interpretations of these stories (Anderson et al., 2014; Brossard and Scheufele, 2013). In this noisy information landscape, scientists have difficulty finding responsible ways to ensure that the public has access to clear and credible evidence.

Opportunities for Communicating Science: Social Media, Social Networks, and Blogs

Social Media

Social media offer expanded opportunities for science communication and the exchange of ideas, but they differ in important ways from traditional media and even online versions of traditional media outlets. Social media offer platforms with user-generated content and interactions, and their information is tailored and targeted toward specific individual users (Cacciatore et al., 2016). The platforms vary in the extent to which they enable people to have a social presence, and people vary in how much information they disclose about themselves (Brossard, 2012). Generally, however, like-minded individuals can find one another and interact through social media. They can form social networks and communities that share and spread information with great speed. (This effect is enhanced when the information flows through particularly influential individuals within a network [Brossard, 2012].)

The popularity of a given platform or site and the demographic characteristics of its users often shift rapidly (Kümpel et al., 2015). As of this writing (in 2016), young adults, for example, are increasingly using Snap-chat and Instagram, while their use of Facebook is decreasing (Brossard, 2016). Partly as a result of this highly dynamic environment, the effects of a particular platform are difficult to disentangle from other influences that may explain an individual’s attitude, preference, or behavior. These realities pose challenges for researchers seeking to understand the impact of social media on science communication. A recent review of research conducted between 2004 and 2014 on news sharing in social media (Kümpel et al., 2015) identified a number of important tasks for researchers, including the need to keep pace with changes in the media landscape, the need to devise more comprehensive models of news and social media, and the need to derive more information about news-sharing networks.

Scientists’ use of social media An emerging body of research examines scientists’ own use of social media. Some researchers have found that scientists’ use of Twitter can amplify their scientific impact, with tweeted work being cited by other scholars more frequently than other work (Eysenbach, and Köhler, 2002; Liang et al., 2014). Other research from the biomedical and health domains contradicts these findings, however, with articles shared on Twitter and other social media being no more likely than nonshared articles to be cited by other researchers or downloaded (Haustein et al., 2014; Tonia et al., 2016). Scientists’ impact scores also have been found to improve with increased interaction with reporters being mentioned on Twitter, and interaction with members of the public if it is mentioned on Twitter (Liang et al., 2014). Others have noted, however, that the science community is slow to change the way it communicates, for a variety of reasons, even though scientists see the value in communicating the work they do (Peters, 2012b, 2013).

Organized uses of social media According to a review of research focused on climate change communication online and via social media (Shafer, 2012), social media are important for the strategic communication of a variety of stakeholders, although substantial gaps in the research in this area exist. In the research reviewed, however, climate scientists and scientific institutions played a minimal role in online debates, and their activity was limited primarily to blog writing. Nongovernmental organizations are much more prominent online communicators, but less is known about the climate communications of other corporate and government stakeholders. Research is beginning to examine the impact of government science communication online as well, particularly in response to crises (e.g., Ebola and lead contamination in Flint, Michigan) (Dalrymple et al., 2016).

Individuals and organizations can use social media to facilitate the spread of unverified rumors and myths (Del Vicario et al., 2016). Further, those who use these channels to communicate about science-related controversies can disproportionately be people with views counter to the scientific consensus (McKeever et al., 2016). Other research has found that the communication on social media also is prone to the “spiral of silence,” the phenomenon whereby individuals are hesitant to speak out if they feel that a majority of other people do not share their views (Hampton et al., 2014).

Social Networks

Social networks long predate web-based platforms such as Facebook and Twitter. Social networks generally are the web of connections and

relationships people have with others—connections that vary in strength (Kümpel et al., 2015) and provide a means for social influence (Contractor and DeChurch, 2014). The importance of networks in policy systems is discussed in Chapter 3. But given that all people exchange information through such connections, social networks may offer science communicators a means of reaching audiences who do not follow journalistic media or use web-based platforms. These underserved audiences (generally less educated and less affluent) can potentially be served by the astute use of their social networks to transmit accurate and useful scientific information and to counteract falsehoods and distortions.

The spread of beliefs, attitudes, and behaviors through networks has been observed in the areas of obesity, alcohol consumption, and smoking, as well as political mobilization and cooperation (Christakis and Fowler, 2013). Researchers have used social network analysis to determine statistically how these networks exert influence on individual behaviors. This work remains the subject of debate. Some have suggested that the noted effects, attributed to the influence of social networks, may have other causes, such as homophily, or the tendency of people who have common interests, traits, and characteristics to make similar decisions (Aral et al., 2009; Noel and Nyhan, 2011; Shalizi and Thomas, 2011).

The influence of opinion leaders in social networks is another means by which these networks affect individuals, and has been tapped in efforts to disseminate information from science to improve public health (Contractor and DeChurch, 2014). Such approaches hold promise for communicating scientific information effectively to improve the impact of science on public welfare. One study found that nonscientists are more likely to share scientific information with others in their networks when it arouses emotion, or seems useful or interesting to lay audiences, for example (Milkman and Berger, 2014).

With one in seven people worldwide regularly using digital communication platforms such as Facebook, an empirical understanding of how science-related content is communicated within and across social networks is crucial. This understanding may be particularly useful for reaching audiences who are less likely to engage with science-related content in mass media or other traditional channels. Thus online social networks afford a number of opportunities for future research. More systematic analysis is needed of how people understand and perceive science through social media and their social networks. Such tools as social network analysis (Watts and Dodds, 2007) could be used to document the flow of science-related information and sentiments in social networks (Runge et al., 2013; Su et al., 2016) and assess its effects.

Blogs

Science-related blogs have served as one way for highly motivated segments of the public to learn about, follow, and discuss science. These blogs blend the textual depth of online newspapers with the graphical and video capabilities of television, and they enable readers to interact in real time with the blog’s author. Posts also can be written quickly and immediately, responding to new events, issues, or debates, and bypassing the need to convince a journalist to write about or an editor to publish an op-ed on the topic.

Still, data on broader public consumption of science-related blogs are limited, and those studies that do exist suggest that blog reading occurs among a small, unique segment of the public. This population is not necessarily drawn to science blogs for their discussion of science. Many individuals who seek out science blogs may be looking for discussion of the politics of such topics as evolution, atheism, and climate change instead of wishing to learn about science more generally (Su et al., 2014).

Then again, science blogs may be more likely to reach scientists, funders, decision makers, and journalists than the general population. A 2009 Pew Research Center survey of members of the American Association for the Advancement of Science about the state of science and its impact on society found that although only 9 percent of respondents wrote a science blog, 42 percent read such a blog very often or occasionally (Pew Research Center, 2009a). Other research suggests that journalists often use science blogs as a source of story ideas or to track specialized areas of research (Fahy and Nisbet, 2011). In other research, positive relationships were observed among news coverage, social media mentions, and a scientist’s total citation impact scores (Liang et al., 2014).

EMERGING RESEARCH ON USE OF THE INTERNET AS A SOURCE OF SCIENCE NEWS

Research on how changing online news environments have influenced the way scientific information is communicated is still in its infancy. Initial studies suggest that the promise of more easily available scientific information does not necessarily translate into more effective science communication (Brossard and Scheufele, 2013). The emerging research on the use of the Internet as a source of science news can be grouped into at least three broad areas: preference-based effects, contextualized news, and widening knowledge gaps.

Preference-Based Effects

Preference-based models of media effects focus on the use of data on people’s behavior (by media organizations or by audiences themselves) to tailor information to the preferences of different audiences (Cacciatore et al., 2016). In online communication environments, information is increasingly tailored to specific audience members based on data on location, prior search history, click rates, shopping behavior, and a host of other personal information (Lazer, 2015; Pariser, 2012). While much attention has been focused on concern about algorithms affecting social media news feeds, even traditional news outlets rely on algorithms to make decisions about news selection or placement, a phenomenon that has been termed “algorithms as editors” (Peters, 2010). Tailored or narrowcast online messaging may be more likely to reach audiences than messaging in traditional broadcast models because, relative to traditional media, online environments make it easier for audiences to select news sources consistent with their beliefs and preferences (Yeo et al., 2015). As a result, people are more likely to pay attention to these messages once exposed to them (Cacciatore et al., 2016; see the discussion of motivated reasoning in Chapter 3). Some empirical work combining search histories from Nielsen NetRatings, Google recommendations, Google search results, and analysis of web content, for example, has shown that the way popular search engines such as Google present search results affects the scientific information people are likely to encounter (Brossard and Scheufele, 2013). In addition to concerns that algorithms may provide only information that consumers find agreeable, algorithms may lead people to what is popular rather than to what is most accurate and useful (Downs et al., 2008; Ladwig et al., 2010). These feedback loops limit the diversity and quality of the information people encounter when they search, a phenomenon that is particularly disconcerting given the above-noted prevalence of search engines as a source of information about science (Brossard and Scheufele, 2013).

In addition to the tailoring of news for particular audiences, online channels, such as news aggregators and social media, give audience more efficient tools than they had with traditional media, such as television and newspapers, for preselecting the types of content to which they would like to be exposed (Scheufele and Nisbet, 2002). A recent study examining public reactions to silver nanoparticles, for instance, offered respondents the opportunity to select from a given set of background articles on the topic before answering more questions. Each article was randomly assigned an ideological cue about the source (conservative, liberal, or neutral), similar to what would be provided by a news aggregator such as Feedly, Flipboard, or Google News. In line with the concept of preference-based selection, con-

servative respondents in one study were more likely to select Fox News as their first information source, regardless of content, while liberals were significantly more likely to select MSNBC (Yeo et al., 2015). In other words, people’s tendency to attend disproportionately to news content coming from ideologically consistent sources is further enabled by new information environments that allow them to make this selection with limited effort, and before even seeing any content with which they might disagree.

Contextualized News

Research has examined the influences of social feedback on how science news is used. Research in communication science, social psychology, and related fields has shown that people’s perceptions of social norms (i.e., their sense of appropriate or expected behaviors and attitudes) shape their own behaviors and attitudes (Noelle-Neumann, 1993; Schultz et al., 2007; Sherif, 1967). Online news environments have allowed for almost instantaneous social feedback through Facebook likes, YouTube views, reader comments, Reddit upvotes, and retweets on Twitter. An emerging body of research suggests that this social contextualization of news can have a significant effect on how audiences view and process science news. Work on what has been called the “nasty effect,” for instance, has shown that being exposed to uncivil reader comments on objective scientific reporting on a topic can increase readers’ perceptions of bias in the story itself and can even polarize perceptions of the risk associated with the topic of the story (Anderson et al., 2014; Brossard, 2013; Brossard and Scheufele, 2013).

Cues that signal the accepted social norms related to a topic can be much more subtle than explicitly uncivil comments. The simple number of views for YouTube videos is an example of such a cue. A recent study, for instance, linked the number of views listed under YouTube videos about climate to perceptions of how “others” feel about the climate issue. The number of views cue did more than influence perceptions of other Americans’ thinking on the importance of climate change as an issue. Among those respondents who were generally more responsive to social cues, this cue increased their own judgments about the issue’s importance (Spartz et al., 2015b).

Widening Knowledge Gaps

Evidence indicates a persistent knowledge gap with respect to new media between less-advantaged people and those with higher socioeconomic status and more education (Cacciatore et al., 2014; Eveland and Scheufele, 2000; Slater et al., 2009; Tichenor et al., 1970). This knowledge gap reinforces status quo differences in resources and participation in society’s

conversations about contentious issues and the related science. Science communicators need to consider this gap in planning how to engage segments of the public not already attending to or interested in science as a topic. People who visit museums, watch documentaries such as Nova Science Now on the Public Broadcasting Network, or attentively read the science coverage of a national newspaper are an easy audience to reach. But they are not the majority of the American population, and in education and wealth, they represent a kind of elite.

Research has documented that differences in how scientific information is used and processed can, in fact, widen gaps in knowledge among groups. In particular, groups with higher socioeconomic status process and absorb new information more efficiently than groups with lower socioeconomic status. As a result, existing gaps in knowledge widen as new information becomes available, since audiences with higher socioeconomic status also tend to have more informed discussion networks to help them make sense of new information or bring knowledge that allows them to process new information more efficiently (Corley and Scheufele, 2010; Nyhan et al., 2013; Scheufele, 2013). Early research also suggests, however, that Internet-based sources of scientific information can be more accessible to diverse audiences regardless of socioeconomic status and help narrow knowledge gaps produced by more traditional outlets (Cacciatore et al., 2014).

Research is needed to understand how individuals and decision-making bodies derive and evaluate information from various media sources. Research also is needed to determine how science communicators can be heard among many competing sources of information, messages, and frames. Further, are some forms of media better than others in promoting awareness or understanding of or informing public opinion about science and scientific information, and for whom?

This page intentionally left blank.